将其转换为时间序列

Question

我是R.的新手已经采用了泵的每日数据。这些数据为两年，共有742个数字。

library(tidyverse)

sig_breaks <- c(zero = 0, "***" = 0.001, "**" = 0.01, "*" = 0.05, NS = Inf)

iris %>%
  as_tibble() %>%
  select_if(is.numeric) %>%
  gather(key = measure, value = value, -Petal.Width) %>%
  group_by(measure) %>%
  do(mtx = cor.test(.$value, .$Petal.Width, method = "spearman")) %>%
  broom::tidy(mtx) %>%
  mutate(stars = cut(p.value, breaks = sig_breaks, include.lowest = T, labels = names(sig_breaks)[2:5]))
#> # A tibble: 3 x 7
#> # Groups:   measure [3]
#>   measure      estimate statistic  p.value method        alternative stars
#>   <chr>           <dbl>     <dbl>    <dbl> <fct>         <fct>       <fct>
#> 1 Petal.Length    0.938    35061. 8.16e-70 Spearman's r… two.sided   ***  
#> 2 Sepal.Length    0.834    93208. 4.19e-40 Spearman's r… two.sided   ***  
#> 3 Sepal.Width    -0.289   725048. 3.34e- 4 Spearman's r… two.sided   ***

iris %>%
  select_if(is.numeric) %>%
  as.matrix() %>%
  Hmisc::rcorr(type = "spearman") %>%
  broom::tidy() %>%
  filter(column1 == "Petal.Width" | column2 == "Petal.Width") %>%
  mutate(stars = cut(p.value, breaks = sig_breaks, include.lowest = T, labels = names(sig_breaks)[2:5]))
#>        column1     column2   estimate   n      p.value stars
#> 1 Sepal.Length Petal.Width  0.8342888 150 0.0000000000   ***
#> 2  Sepal.Width Petal.Width -0.2890317 150 0.0003342981   ***
#> 3 Petal.Length Petal.Width  0.9376668 150 0.0000000000   ***

将其转换为时间序列

cw<-read.csv("RCW1.csv")

str(cw)

data.frame':    742 obs. of  14 variables:
 $ date       : Date, format: "2016-04-01" "2016-04-02" "2016-04-03" ...
 $ amp        : num  226 227 0 225 226 ...
 $ brg_de_tmp : num  38.1 38.1 39.6 41.6 41.5 ...
 $ brg_nde_tmp: num  78.6 79.1 72 79.9 80.4 ...
 $ kg         : num  2.07 2.07 0.06 2.29 2.28 2.3 2.11 2.1 2.11 2.11 ...
 $ level1     : num  8.45 8.46 8.69 8.67 8.43 8.6 8.39 8.5 8.46 8.65 ...
 $ level2     : num  8.44 8.46 8.67 8.65 8.42 8.59 8.38 8.48 8.46 8.63 ...
 $ mde_xvib   : num  1.15 1.35 0.28 1.05 1.15 1.06 1.25 1.25 1.25 1.25 ...
 $ mde_zvib   : num  1.37 1.57 0.4 1.18 1.13 1.38 1.28 1.57 1.3 1.5 ...
 $ rpm        : num  296.46 296.91 -4.76 297.09 297.91 ...
 $ mde_yvib   : num  2.09 2.38 0.34 2 1.82 2.24 2.17 2.56 1.9 2.27 ...
 $ m_nde_yvib : num  1.15 1.13 0.35 0.96 0.96 0.96 1.15 1.06 1.15 1.15 ...
 $ m_nde_zvib : num  1.53 1.63 0.27 1.33 1.43 1.4 1.76 1.63 1.79 1.71 ...
 $ permit     : chr  "#N/A" "#N/A" "CW Pump house: Motor stand" "#N/A" ...

我有两个问题。

编号1：如何在绘制以下代码时在X轴上获得“日期”。

cw_x <- xts(cw, order.by=as.Date(cw[,1], "%Y/%m/%d"))

cw_ts<-as.ts(cw_x)

> head(cw_ts)
Time Series:
Start = 1 
End = 6 
Frequency = 1 
        date    amp brg_de_tmp brg_nde_tmp   kg level1 level2 mde_xvib mde_zvib    rpm mde_yvib
1 2016-04-01 226.05      38.06       78.61 2.07   8.45   8.44     1.15     1.37 296.46     2.09
2 2016-04-02 226.59      38.08       79.13 2.07   8.46   8.46     1.35     1.57 296.91     2.38
3 2016-04-03   0.00      39.57       71.96 0.06   8.69   8.67     0.28     0.40  -4.76     0.34
4 2016-04-04 225.01      41.57       79.91 2.29   8.67   8.65     1.05     1.18 297.09     2.00
5 2016-04-05 226.41      41.54       80.43 2.28   8.43   8.42     1.15     1.13 297.91     1.82
6 2016-04-06 225.65      41.08       79.89 2.30   8.60   8.59     1.06     1.38 297.55     2.24
  m_nde_yvib m_nde_zvib                     permit
1       1.15       1.53                       #N/A
2       1.13       1.63                       #N/A
3       0.35       0.27 CW Pump house: Motor stand
4       0.96       1.33                       #N/A
5       0.96       1.43                       #N/A
6       0.96       1.40                       #N/A

请参阅我在X轴上获取行号，我想更改为日期。

问题2：我正试图通过以下两个代码获得季节性情节，如Robjhyndman的书“预测：原则与实践”所述

autoplot(cw_ts[,2:5],facets = TRUE) + ylab("parameters")

和

ggseasonplot(cw_ts, year.labels=TRUE, year.labels.left=TRUE) +
ylab("") +
ggtitle("Seasonal plot: Pump parameter")

那么如何将我的时间序列归入上面的代码中呢？我收到了以下错误。

ggseasonplot(cw_ts, polar=TRUE) +
ylab("") +
ggtitle("Polar seasonal plot: Pump parameter")

和

ggseasonplot(cw_ts, year.labels=TRUE, year.labels.left=TRUE) +
+   ylab("") +
+   ggtitle("Seasonal plot: Pump parameter")
Error in ggseasonplot(cw_ts, year.labels = TRUE, year.labels.left = TRUE) : 
  Data are not seasonal

任何建议都会对我有所帮助。谢谢。

每日数据的ggseasonalplot不起作用

将其转换为时间序列

0 个答案: